While computing devices are capable of performing increasingly complex calculations and functions, attempts at configuring a computing device to perform a truly random operation have largely failed. This is because, at a fundamental level, computing devices operate by following instructions to complete tasks. Due to their deterministic nature, computing devices struggle to generate truly random numbers, and historical tools for “random” number generation have been pseudo-random, insofar as they utilize methods and instructions that can be reproduced if the inputted data and transformations are known. In this way, computing devices attempt to mimic true statistical randomness but do not actually achieve it.
In some instances, pseudo-random numbers are used as a means for securing or otherwise authenticating a session or other interaction between two devices by serving as a unique session identifier. Traditional methods for generating pseudo-random numbers for authenticating a session, however, have become increasingly susceptible to attack as the availability of computing power has increased. Perpetrators now may utilize this increased computing power, along with access to a user's device or information, to exploit the vulnerability of traditional pseudo-random number generation techniques. Furthermore, due to the current prominence of data collection and the associated quantity of information available about any particular user, pseudo-random numbers generated based upon a user's personal information face an increased risk due to the increased likelihood that the personal information may be known by an attacker.